The present invention relates to an image reader that is appropriate for a TV telephone or a TV conference system, and relates in particular to an image reader that can be compactly and lightly constructed, and that can accurately and quickly read the image of even a distorted object.
An example, conventional image reader is disclosed in JP-A-11-136564.
This image reader comprises: an image pickup unit, for obtaining, using an image forming optical system having a variable magnification rate, an image of an object document at a predetermined magnification rate; a distance sensor, for measuring distances to three targeted points on the object document; and an image synthesizer, for employing, when the image pickup unit sequentially acquires the images of adjacent targeted portions at the same magnification rate, the distance values obtained by the distance sensor to transform multiple, thus obtained images into those equivalent to images viewed from the front, and for synthesizing the transformed images to prepare a single image of the entire document. According to this image reader, the document can be read very accurately, even when the document is positioned so that it is tilted. Further, since glass platen that is equal to or larger than a document is not required, the size of the image reader can be reduced.
However, since a premise of the employment of a conventional image reader is that a document have a flat face, such an image reader can not satisfactorily read a document that is bent, and distorted document information is obtained. When, for example, a page of an opened book is to be read, the images, such as character images, at the bound portion, where the page is bent, can not be read accurately.
When a document that is tilted is to be read, the distance to the image pickup unit varies, depending on the portion of the document that is to be read. Thus, the sizes of the individual images differ, and the synthesized image is distorted. In this case, the magnification rate can be changed by referring to the characteristics of adjacent images for the performance of the synthesization process. However, according to this method, the magnification rate must be changed repeatedly, and an extended time is required for the synthesization process.